EVTV celebrates eight years of publication this month. Our first video was published May 22, 2009 and we have just uploaded video number 312 - two hours, nineteen minutes, 50 seconds. When we began this journey, there were ZERO electric cars available for purchase anywhere in the world - Tesla Roadster being much talked about but unobtainium at the time. We've averaged 39 videos per year. GAV's EV series being at an end, we were the only savage on the buffalo hunt and indeed I was early advised that no one would have any interest in a two hour video but in any event no one could view it in what was then considered high definition of 1280x720. The bandwidth didn't support it but many older PC's wouldn't even play it.
Today it is rare for anyone to convert a vehicle to electric drive without doing a video series on it and 4k is the new high water mark in resolution. There are today essentially ZERO OEM's who do not offer some version of an electric vehicle. In many ways, we have outlived our purpose and become an anachronism. And we've managed the entire cycle with sufficient ineptitude to have never actually suffered a profit financially.
We have always hosted the videos on Amazon's AWS cloud service and less than 10% of EVTV viewers watch it on the ever annoying YouTube. But for those watching a two hour video on a cell phone, they still prefer it there. And actually for some months it has been easier to include the Youtube version in the blog than to deal with the vagaries of a now very aging version of JW Player.
ALong the way, an odd thing has happened. While our very first EV, the 1967 Porsche Speedster, is here in the front window of the shop and still runs, we do less conversions as time progresses and maintain the others poorly. The Cadillac Escalade gave up its transmission for the second time and I just have no interest in dealing further with automatic transmissions. They have proven a poor choice for electric vehicle use. I still love the Escalade, and we may rebuild it with a Tesla rear end or something. But as stated in an earlier blog, I am feeling overwhelmed by a target rich environment.
And the target seems ever further removed from cars. OEM cars have come online with very advanced drive and battery systems, making our early efforts with fork lift motors and basic PWM controllers seem quaint at this point. Most of the players in the 2008/2009 space are long gone. And we've moved more toward adapting these OEM components for use in electric car and boat builds.
This has had the effect of sucking us down a wormhole of every more technical investigation into CAN and circuit board design and software development, and ever further from actually building a car.
The attraction of course is vastly better components and engineering ever more available from a vast pile of wrecked low-mile manufactured electric vehicles. But the manufacturers treat every aspect of their operation as some sort of sacred state secret, and so we have to dig harder and harder to make use of them. And as such their utility is dramatically limited. A battery or motor or controller is readily available because nobody knows how to make it work, the market for replacement/repair parts is ridiculously limited, and so they have essentially zero value. Until or unless someone makes them actually do something.
That Tesla would invest over $20,000 in each drive unit, and in the end, with 10,000 miles on it and in perfect operating condition, it is too often scrapped for the weight and value of the metals in it seems vaguely criminal.
But everyone falls into the game.
Speedhut has announced a series of customizable dash gages that use the J1979 OBDII protocol for light passenger vehicles. Yes, in the video I refer to it as J1939 repeatedly and in error. J1939 is similar but for trucks and heavy equipment. Take a look at their TOTAL documentation for these gages. Two pages that describe NOTHING of its operation. It is laughable. They assume you are going to plug these into an existing car through the OBDII port. If you have an existing car with ECU and OBDII port, doesn't it already HAVE gages? But their product offering is heavily skewed to that scenario. And they provide actually ZERO information on how they work. And using standard published specification PIDS, kind of imply it's a state secret.
IN a separate error I noted that they don't actually comply with J1979 as they don't send PID requests. Their technical guy insisted they DID send PID requests. Turns out it was kind of a point of view thing.
The gages actually send out ONE PID request. That is for PID 0. PID 0 is a request for a list of which of the first 32 standard PID codes is supported by the vehicle ECU. The gages send this 0x7DF request ONE TIME IN A ROW. And if it doesn't receive the list, it assumes it is NOT connected to a viable CAN bus, and never sends anything at all again. If it DOES receive the reply message, it happily starts spewing PID requests for data in a fairly vigorous fashion.
There is a lot wrong with this. CAN is a more noise resistant serial network but it is still operating in a noisy environment. So there is no assurance that a single message will get through. So every device on the bus tends to blast CAN messages all over the place in kind of a broadcast mode hoping something will get through. And a very crude bus arbitration scheme settle all disputes in favor of the LOWER number message IDs. 0x7DF is one of the HIGHEST or LEAST PRIORITY messages you can send, 0x7EF and 0x7FF being the only thing it has priority over.
Unfortunately, the gages come up and issue this request before our controller gets through its setup routines. So by the time we are ready to respond, the gages have already shut down, never to try again. Unless we recycle power to them of course.
Actually Speedhut has been very helpful on the phone and using Collin's SavvyCAN we have worked out just how they work and how we can incorporate this with the Tesla Drive Unit. But in writing the seven page document addendum to our ALSET controller kit, it occurs to me that this is really pretty involved on a very technical level to get 5 gages to display Speed, RPM, Voltage, Temperature and state of charge. And it appears I must write the documentation for their product, with their technical representative a little unschooled on just how the things DO work?
It is a herd mentality run amuck. I'm reminded of when MacDonalds introduced their breakfast menu, but had a sharp turnover from breakfast to hamburgers at precisely 10:30 each morning. There were of course operational reasons for doing so. But it doesn't explain why restaurants and fast food places across the entire country suddenly decided that they would end breakfast promptly at 10:30 when previously they would of course serve you breakfast at any time of the day or night. But within days of MacDonalds announcement, you couldn't buy an egg for love or money after 10:29.29 AM anywhere.
I actually went into our restaurant at the Cape Girardeau airport one morning about 10:40. This poor girl was trying to make a living running a restaurant at an airport that had a few airplanes, but no people there most of the time. There wouldn't be 8 guys at the airport most days all day. I ordered breakfast and she looked at me and very firmly informed me they didn't serve breakfast after 10:30. I looked around the restaurant at all the other tables and chairs in amazement. She and I were the only people IN THE BUILDING. And she was chanting to me some nonsense about being 10 minutes too late to order an egg WHILE she is slowly going bankrupt in front of me. And there's not even another humanoid in the place to witness this. But she had gotten this from a MacDonalds operation that had entirely different reasons for the policy. So it made perfect sense to her.
It is no secret I'm a huge Tesla fanboyz and indeed, at this point I think they are so far ahead of the game that General Motors, Ford, and BMW, Volkswagen et al really don't even matter. They have so much blue sky opportunity in front of them, with an 85 billion automotive market before them, offering totally disruptive change, that it is a huge technology party.
I was touched by Elon Musk during a very recent TedTalk interview. You have to kind of wade through to the end at 39:40 to hear it. The interviewer is sucking and stroking so hard it is even embarrassing for Musk. He finally states "I'm not trying to be anyone's savior....I just want to find ways to think about the future without being sad." Probably the best Elon Musk quote of all time and I think very truthful.
The closest thing to criticism came with an actual accusation that he was somehow betraying the faith TALKING to President Trump about climate change. This was pathetic. He is kind of dependent on government for many of his SpaceX plans, so certainly doesn't need any enemies in the white house. And if he doesn't show up to talk about it, how is Trump ever going to hear it?
I've actually heard Musk on the topic before. He is truly a bright guy and I would claim him in my camp. X=don't know. Climate is a much more complex and variable system than most of the denialist deniers (denialist squared???)quite grasp. But Musk is a risk/benefit statistical probability guy. The risks are high - we have one planet and if we screw it up we're hosed. And the benefit very low - we keep using fossil fuels and what? As Shah Agassi said, we didn't leave the stone age for lack of stones. What's the benefit to driving around like Jed CLampett in an ancient oil burner with Granny tied on back in a rocking chair?
So you factor it out. Put global warming on both sides of the equation and it cancels out. Leaving you with the solution - electric cars and solar energy. Why are you rooting through the temperature data and arguing about it so much? With a clear solution available, why all the heat and stress over what the problem really is? The solution carries so many OTHER benefits it really doesn't matter. We should do it. We are going to do it. And we're working on doing it now.
In any event, Tesla's stock is now trading for ten times its value shortly after IPO. And I think we are looking at another 10x in valuation over the next five years. Tesla's actual financial performance actually doesn't matter. They have the ability to raise unlimited funds. And they are a technology growth company, not an automobile manufacturer.
Back to earth here at EVTV, we are in an ever tightening downward spiral going ever deeper in the technical aspects of Inverters and motors and batteries and more pointedly, the electronics used to control them and our designs of software and hardware to control THAT. I have a drive unit test bench, a charger test bench, and two battery test benches all mostly about laptop computers, CAN serial communications, and how to simulate a 20ma current flow into a single obscure connector pin to allow full power out of the drive unit. That's where we were this week.
And I'm writing software to display CAN message bit fields that can't actually BE reverse engineered - each bit representing some random made up error fault code. But there are up to 64 different errors represented in one CAN message, repeated and updated every 100 milliseconds.
All this to say that the history of automobiles and the rich tradition of reusing used parts and third party products and so forth is basically being nullified by proprietary technologies. In examining a later model inverter and comparing it to an early model inverter we discovered Tesla has actually gone to a serious inconvenience for themselves and locked down the code in the inverter firmware using the Code Protection Scheme provided by Texas Instruments. This further complicates the reuse of future Tesla components with no discernible advantage to Tesla whatsoever that I can fathom. It just IS 10:30 and I'm sorry sir. No breakfast.
That is not to say that it is some vast Tesla conspiracy to end automotive repair. As I told Collin Kidder and Jarrod Tuma, who are working on these Inverter controller boards, there is no Tesla. It is now a large organization, a corporation. And within Tesla, there aren't four guys with the knowledge and ability to do what you two are doing, or even to know what you are talking about. But a game has been launched by the self aggrandizing self promoting wannabee "hackers" to find holes in the Tesla system. And Tesla now has people dedicated to plugging the holes. It's a bit mindless, if damaging to our efforts. They innovate continuously. And we have to do the same. It is what it is.
But it is expensive. And it has taken us far afield from our original raison d'etre in an ever tightening spiral of ever more specialized electronic and software work. We've come a long way from picking up for GAV's EV. I don't see any future where this focus again widens. And it will be interesting to see where this wormhole spits us out in the future.
The DOKA is nearing completion. Bill had a wiring harness in. I don't know if the axles will hold, and I don't know our little Influenza battery will hold up if they do. But we should be rolling in a week or so with a Tesla Drive Unit powering it.
There were some design flaws in our first prototype run of Tesla battery module controller boards. We have had to go for a second round of battery module interface boards for the Tesla Battery Module Controller so that will be another 3 weeks delay. And so it goes...
Jack Rickard
